Women with epilepsy initiating a progestin IUD: A prospective pilot study of safety and acceptability

Transcription

1 FULL-LENGTH ORIGINAL RESEARCH Women with epilepsy initiating a progestin IUD: A prospective pilot study of safety and acceptability *Anne R. Davis, Heva J. Saadatmand, and Alison Pack SUMMARY Dr. Anne Davis is an associate professor of obstetrics and gynecology at Columbia University Medical Center. Objective: Effective contraception enables women with epilepsy (WWE) to plan their pregnancies and improve outcomes for themselves and their children. Although popular among all women, complex drug interactions limit the efficacy and safety of oral contraceptives (OCs) for WWE. We sought to explore the safety, acceptability, and pharmacokinetic impact of a progestin-containing intrauterine device (IUD) in WWE. Methods: We enrolled 20 women with well-controlled epilepsy and a stable antiepileptic drug (AED) regimen and who were initiating a progestin-containing IUD (levonorgestrel 52 mg) in a prospective, observational study. For each AED, we compared the trough concentration before IUD insertion to the trough concentration 3 weeks, and 3 and 6 months later. Participants recorded seizures in a daily paper diary. We compared seizures that occurred during the month before IUD insertion to those occurring in the 6 months thereafter. Participants completed an acceptability questionnaire at 3 and 6 months. Results: Participants average age was 28 years; 60% were nulligravid. They reported a history of multiple seizure types. During the baseline month, 75% were seizure-free and the remainder reported between one and three seizures. Fourteen received monotherapy and six received polytherapy. Lamotrigine use was most common (n = 12). AED trough concentrations remained stable during the 6 months after IUD insertion, without clinically meaningful deviations from baseline. Diary data showed that seizure frequency worsened in 3, and remained unchanged in 13 and improved in 4 after IUD insertion. Subjectively, no participant believed the IUD worsened her seizure control. All participants were either somewhat or very satisfied with the IUD throughout the study. All participants continued the IUD use at 6 months. No pregnancies occurred. Significance: This pilot study suggests that the progestin-containing IUD is a safe and acceptable long-acting contraceptive for WWE. KEY WORDS: Epilepsy/seizures, Antiepileptic drugs, Women, Contraception, Intrauterine device. Contraception is an essential health concern for women with epilepsy (WWE). 1 When WWE seek pregnancy, effective contraception allows healthcare providers to choose Accepted August 12, 2016; Early View publication 28 September *Department of Obstetrics and Gynecology, Columbia University Medical Center, New York, New York, U.S.A.; Columbia University College of Physicians and Surgeons, New York, New York, U.S.A.; and Department of Neurology, Columbia University Medical Center, New York, New York, U.S.A. Address correspondence to Alison Pack, 710 West 168th Street, New York, NY 10032, U.S.A. ap390@cumc.columbia.edu Wiley Periodicals, Inc International League Against Epilepsy antiepileptic drugs (AEDs) with lower teratogenic risk, initiate supplemental folic acid, and optimize seizure control before conception. Oral contraceptives (OCs) are commonly prescribed for young women; however, OC use in WWE is challenging. Coadministered hepatic-enzyme inducing AEDs reduce OC efficacy, 2 and estrogen-containing OCs cause estrogen-mediated decreases in lamotrigine concentrations. 3 The intrauterine device (IUD) provides long-acting, highly effective, and reversible contraception. Available IUDs contain copper or the progestin levonorgestrel. For 1843

2 1844 A. Davis et al. Key Points This small, preliminary study found that no clinically meaningful changes in seizure control or AED trough concentrations occurred during follow-up compared to pre-iud insertion Most WWE choosing the progestin-containing IUD were treated with lamotrigine No pregnancies or serious adverse events occurred These WWE found the progestin-containing IUD to be acceptable, with all continuing use at 6 months some women, the progestin IUD is preferred because decreased menstrual flow is expected. Drug interactions with hormonal methods and AEDs are clinically important. Limited published evidence, however, addresses the impact of a hormonal IUD on AED concentrations, seizure control, and acceptability. 4,5 In this observational, prospective pilot study, our objective was to characterize the safety of a progestin-containing IUD by measuring AED concentrations and seizures in WWE before and after initiating the IUD for contraception. We also assessed IUD acceptability by measuring continuation at 6 months and satisfaction using a questionnaire. Methods The Columbia University Medical Center Institutional Review Board approved this study. Participants provided written informed consent prior to any study procedures. We conducted a prospective cohort study of 20 WWE initiating a progestin-containing IUD from August 2011 to October 2013, at an urban academic medical center. Our inclusion criteria were age years, regular menstrual cycles between 21 and 35 days in length, seeking contraception, confirmed diagnosis of epilepsy, stable AED regimen (no change in dose or type 2 months before enrollment), and well-controlled epilepsy ( 2 seizures per month, excluding focal seizures with no impairment of awareness). We excluded women with more pharmacoresistant epilepsy in order to isolate effects of the progestin IUD on seizure frequency. Exclusion criteria were short-acting hormonal contraceptive use in the month prior or depomedroxyprogesterone acetate in 6 months prior, IUD contraindications, pregnancy in the last 2 months, and irregular menstrual cycles. We included nulliparous women. A baseline questionnaire recorded demographic characteristics, reproductive and contraceptive history, medication use, and seizure history. Participants provided a description of seizure(s). Seizures were then categorized by the study neurologist (AP) and clarified with the treating neurologists. We did not review imaging or electroencephalography (EEG) findings to establish seizure type. Each participant s neurologist also confirmed eligibility and AED type and dose. After the enrollment visit, participants began recording seizure type and occurrence and menstrual bleeding in their diaries. At the next menses, we reviewed the diary and collected serum for AED concentrations, and after a negative pregnancy test result, placed the IUD (levonorgestrel 52 mg, Mirena, Bayer, Whippany, NJ, U.S.A.). We defined the baseline month as the 28 days prior to IUD insertion. For the next 6 months, participants recorded bleeding and seizure type and occurrence in their daily diaries. Participants returned on day 21 of that menstrual cycle and at 3 and 6 months later. At each visit, we reviewed their diaries and medications and collected serum. Data from the diaries will be presented elsewhere. We centrifuged and aliquoted serum for storage ( 70 C) and batch analysis. A blood sample for the first AED trough concentration was collected immediately before IUD insertion. We timed the visit and venipuncture to occur before the next morning or evening AED dose, when possible. We scheduled return visits (day 21 of first menstrual cycle after IUD insertion, 3 months later, and 6 months later) for the same time. Quest Diagnostics evaluated AED concentration by immunoassay for valproate and liquid chromatography mass spectrometry (LC/MS/MS) for the other AEDs. We excluded samples after a change in AED type or dose. We also measured progesterone, levonorgestrel, and serum hormone binding globulin (SHBG); those results will be reported separately. At 3 and 6 months after insertion, participants self-administered a brief questionnaire created by the investigators to assess overall satisfaction with the IUD (very satisfied, somewhat satisfied, neutral, dissatisfied, or very dissatisfied), plans for continuation, and perceived impact of the IUD on epilepsy (better, no change, worse, or don t know). We ascertained adverse events with an open-ended question at each visit and by reviewing diaries. We recorded reported medication changes and confirmed them with the participant s neurologist. This was a pilot, exploratory study. Based on our clinical practice and prior study recruitment, we estimated that a study of 20 WWE seeking contraception would be feasible and adequate to provide preliminary data. We used descriptive statistics to characterize our sample. Results We screened 23 women and excluded 3: one for poor seizure control, one who chose a different method of contraception, and one whose IUD could not be inserted due to surgical scarring. Table 1 summarizes participant demographic and reproductive characteristics, AED therapy, seizure type, and epilepsy syndrome. Nineteen used condoms prior to enrollment; one switched from a copper IUD due to heavy bleeding. Most were treated with AED monotherapy

3 1845 Women with Epilepsy and Progestin IUD Table 1. Baseline participant characteristics Characteristic Mean (SD) Age in years 28 (6.2) Body mass index 25.3 (3.5) N (%) Parity 0 14 (70) 1 2 (10) 2 3 (15) 3 1 (5) Race Caucasian 16 (80) Black 1 (5) Hispanic 2 (10) Other 1 (5) Seizure type Generalized convulsive seizure 6 (30) Myoclonic seizure 5 (25) Absence seizure 3 (15) Focal seizure with no impairment 7 (35) Focal seizure with impaired consciousness 8 (40) Focal seizure evolving to bilateral convulsion 7 (35) Epilepsy syndrome Genetic generalized epilepsy syndrome 6 (30) Juvenile myoclonic epilepsy 3 (15) Localization-related epilepsy syndromes 13 (65) Undetermined whether focal or generalized 1 (5) Antiepileptic drugs Monotherapy Lamotrigine 7 (35) Levetiracetam 4 (20) Oxcarbazepine 2 (10) Carbamazepine 1 (5) Polytherapy Lamotrigine, lacosamide 1 (5) Lamotrigine, oxcarbazepine 1 (5) Lamotrigine, levetiracetam 1 (5) Lamotrigine, valproate 1 (5) Lamotrigine, acetazolamide, clobazam 1 (5) Topiramate, valproate 1 (5) Daily dose range in milligrams Lamotrigine Levetiracetam 500 2,000 Oxcarbazepine 900 1,200 Carbamazepine 800 Lacosamide 500 Valproate 500 Clobazam 10 Topiramate 200 (70%), and 6 with polytherapy (30%). Lamotrigine was the most commonly prescribed medication (n = 12); seven used lamotrigine alone and five in combination with another AED. All participants completed the study and none missed a visit. We obtained serum at all planned time points (n = 80). For 19 subjects, concentrations were measured 10 h from the previous dose. In one participant, serum samples used to evaluate AED concentrations (lamotrigine and lacosamide) were drawn approximately 4.5 h from the previous dose. We included these samples because they were drawn at consistent intervals throughout the study. We excluded six samples obtained after a change in AED type or dose. Two participants had medication adjustments for seizures that occurred before IUD insertion. One participant s medications were changed for an unclear reason; she was seizure-free throughout. Another participant who was taking polytherapy (lamotrigine and levetiracetam) had one medication discontinued (levetiracetam) to achieve monotherapy; as there is no interaction between these AEDs we included subsequent lamotrigine concentrations in the analysis. In two participants, both on lamotrigine monotherapy, the dose was increased after the third month because of increased seizures. Figure 1 shows the trough concentrations of lamotrigine, levetiracetam, and 10-hydroxycarbazepine (oxcarbazepine metabolite) on the day of IUD insertion and 21 days, and 3 and 6 months later. The lamotrigine concentration at day 21 was lower than baseline in 9 of 11 participants, with a median percent change of 9% and range 26% to +8%. By 3 months, the median lamotrigine level was 10% higher than baseline (range 7% to +54%), and by 6 months the median lamotrigine level was 7% higher than baseline (range 13% to +27%). Carbamazepine, clobazam, lacosamide, topiramate, and valproate were each used by only one participant. With the exception of carbamazepine, all concentrations remained within 30% of the baseline. For carbamazepine, all concentrations were subtherapeutic (<1.3 µg/ml). Our analysis included any seizures recorded by participants regardless of medication changes. The most common pattern was to be seizure-free throughout the study (n = 13), including during the baseline month. During the 6 months after IUD insertion, the mean monthly seizure count decreased in four participants and increased in three compared to the baseline month. Of those with decreased mean monthly seizure counts, one participant recorded a worse seizure type after IUD insertion (focal seizure evolving to bilateral convulsion) than during the baseline (focal seizure with impairment ), two had medication changes because of seizures in the baseline period before IUD insertion, and the fourth had no medication changes. For participants who experienced seizures, Table 2 summarizes the seizure timing and type. The four participants with worse control or seizure type were treated with monotherapy (lamotrigine or carbamazepine). The participant treated with carbamazepine had subtherapeutic trough concentrations throughout. Our data showed no relationship between IUD insertion, change in AED trough concentration, and the occurrence or worsening of seizures (Fig. 2). As discussed previously, two of these lamotrigine treated participants had increased doses after the third month. No participants were dissatisfied with the IUD at 3 or 6 months. At 3 months, 50% said they were satisfied and 50% very satisfied. By 6 months, 35% were satisfied and

4 1846 A. Davis et al. endorsed improvement, and two (10%) did not know. No participant endorsed worsening epilepsy because of the IUD. At the final visit, all chose to continue IUD use. No pregnancies, IUD expulsions, or serious adverse events occurred. Twenty-five percent reported uterine cramping. Two (10%) experienced ovarian cysts; in one, the cyst ruptured then resolved without further intervention. Two (10%) reported urinary tract infection, stomachache, muscle spasms, vaginal itching, bacterial vaginosis, or bronchitis. Other adverse events occurred in only one participant. Figure 1. Trough concentrations of lamotrigine, levetiracetam, and 10- hydroxycarbazepine (oxcarbazepine metabolite) on the day of progestin-containing IUD insertion and 21 days and 3 and 6 months later. Data were right-censored after drug discontinuation or dose change. Epilepsia ILAE 65% very satisfied. At 3 months, most participants (85%) endorsed no change in their epilepsy due to the IUD, two (10%) endorsed improvement, and one (5%) did not know. At 6 months, most endorsed no change (85%), one (5%) Discussion In this prospective exploratory pilot study, WWE who initiated contraception with a progestin-containing IUD maintained stable seizure control. Modest changes in AED trough concentrations occurred but did not impact seizure control and did not result in adverse effects during 6 months after IUD insertion. According to data prospectively collected in diaries, 17 of 20 participants experienced no change or an improvement in seizure control after IUD insertion. Three experienced a relative increase in seizure frequency; however, decreases in AED concentrations did not account for the increases. Subjectively, no participant reported that the IUD worsened her epilepsy. Acceptability was high, and all participants continued using the progestincontaining IUD for contraception after the study. Clinicians caring for WWE face challenges when choosing contraception, and a recent U.S. nationwide analysis found lower uptake of effective contraception in WWE compared to women with other chronic medical conditions. 6 Bi-directional drug interactions affect systemic hormonal methods such as combined OCs, the most commonly used method of contraception. The IUD provides a highly effective, long acting, and reversible alternative. OC efficacy is affected by enzyme-inducing AEDs that augment hepatic metabolism. The progestin IUD prevents pregnancy primarily by thickening cervical mucus. This local mechanism of action is unlikely to be influenced by changes in hepatic metabolism; one study found a high efficacy rate among progestin IUD users treated with coadministered enzyme-inducing antiepileptic drugs. 7 Clinicians may be concerned that any hormonal method could negatively affect WWE. The Centers for Disease Control and Prevention s Medical Eligibility Criteria provides guidance for prescribing contraception in women with medical conditions. Their review states that WWE can use the progestin-containing IUD without restrictions, 8 but does not cite evidence regarding seizure control among WWE who use this method. Our results, although preliminary, support that women with well-controlled epilepsy can use the progestin IUD without a clinically meaningful impact on seizures or AED concentrations.

5 1847 Women with Epilepsy and Progestin IUD Table 2. Seizure patterns among participants (7/20) who recorded any seizures during the study Baseline month Six months after progestin IUD insertion ID Seizure number Seizure type Seizure number Mean per month Seizure type 2 3 Focal seizure with no impairment Focal seizure with no impairment 5 1 Focal seizure with no impairment Focal seizure with no impairment 6 2 Focal seizure with no impairment Focal seizure with no impairment 10 0 N/A Focal seizure with no impairment 12 0 N/A One myoclonic; myoclonic, absence, and GCS in 1 day 15 1 Focal seizure with impaired consciousness a Focal seizure evolving to bilateral convulsion 16 1 GCS 0 0 N/A GCS, generalized convulsive seizure. We calculated a mean monthly seizure frequency by combining all seizures that occurred during follow-up and dividing by six. a Although there was a decrease in mean monthly seizure counts, the seizure in follow-up was a worse a seizure type. Figure 2. Timing of seizures and AED concentrations in participants with worsening seizures. Participant ID 6 had an increase in lamotrigine from 400 mg daily to 450 mg daily on study day 157. Participant ID 12 had an increase in lamotrigine from 250 mg daily to 300 mg daily on study day 138. Epilepsia ILAE Lamotrigine is commonly prescribed to reproductive-age WWE, and was the most prescribed AED in our study. 9 For these WWE, contraceptive choice is complicated because the estrogenic components of OCs decrease lamotrigine concentrations substantially, by approximately 50%. 10,11 Estrogen-related changes can be obviated by using progestin-only methods. Published cross-sectional studies support minimal impact of progestin-only methods on

6 1848 A. Davis et al. lamotrigine concentrations. Reimers et al. 5 found comparable lamotrigine dose/concentration ratios in 16 WWE using progestin-only methods (three with an IUD), compared to 18 WWE not using a hormonal method. Among 12 WWE treated with lamotrigine and a progestin IUD, Ohman et al. 4 demonstrated comparable lamotrigine dose/concentration ratios and lamotrigine metabolite (N-2-GLUC)/lamotrigine ratios compared to controls not using a hormonal method. In contrast to these cross-sectional studies, our prospective design allowed us to compare AED concentrations during 6 months of follow-up. We found a median lamotrigine reduction of 9% at day 21; this reduction is substantially less than that reported with combined OCs. This change attenuated by 3 months, and had no meaningful clinical effect on participants seizures. Of the 11 treated with lamotrigine, 6 had no seizures during the study, and 2 improved likely secondary to medication changes in response to seizures before IUD insertion. Of the three who had a worsening of either seizure frequency of type, there is was no relationship to IUD insertion or AED trough concentration changes. Our study has limitations. Each woman served as her own control; a comparison group of WWE unexposed to the IUD would be a more rigorous control group. However, results would be difficult to interpret unless matched on AED and baseline seizure control. Our study was exploratory and limited by the small sample size. We did not assess daily AED compliance, which could account for observed variability in measured AED trough concentrations as well as seizure control. WWE and their children benefit from highly effective contraception and planned pregnancies. Our preliminary results support the progestin-containing IUD as a safe and acceptable long-acting contraceptive for WWE, particularly for WWE treated with lamotrigine. Disclosure of Conflict of Interest Author Dr. Anne Davis received support for this study from Bayer HealthCare Pharmaceuticals, Inc. The remaining authors have no conflicts of interest. We confirm that we have read the Journal s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. References 1. Meador KJ, Penovich P, Baker GA, et al. Antiepileptic drug use in women of childbearing age. Epilepsy Behav 2009;15: Davis AR, Westhoff CL, Stanczyk FZ. Carbamazepine coadministration with an oral contraceptive: effects on steroid pharmacokinetics, ovulation, and bleeding. Epilepsia 2011;52: Sabers A, Ohman I, Christensen J, et al. Oral contraceptives reduce lamotrigine plasma levels. Neurology 2003;61: Ohman I, Luef G, Tomson T. Effects of pregnancy and contraception on lamotrigine disposition: new insights through analysis of lamotrigine metabolites. Seizure 2008;17: Reimers A, Helde G, Brodtkorb E. Ethinyl estradiol, not progestogens, reduces lamotrigine serum concentrations. Epilepsia 2005;46: Champaloux SW, Tepper NK, Curtis KM, et al. Contraceptive use among women with medical conditions in a nationwide privately insured population. Obstet Gynecol 2015;126: Bounds W, Guillebaud J. Observational series on women using the contraceptive Mirena concurrently with anti-epileptic and other enzyme-inducing drugs. J Fam Plann Reprod Health Care 2002;28: Curtis K, Tepper N, Jatlaoui T, et al. US Medical Eligibility Criteria for Contraceptive Use, MMWR Recomm Rep 2016;65: Herzog AG, Mandle HB, Cahill KE, et al. Contraceptive practices of women with epilepsy: findings of the epilepsy birth control registry. Epilepsia 2016;57(4): Sidhu J, Job S, Singh S, et al. The pharmacokinetic and pharmacodynamic consequences of the co-administration of lamotrigine and a combined oral contraceptive in healthy female subjects. Br J Clin Pharmacol 2005;61: Christensen J, Petrenaite V, Atterman J, et al. Oral contraceptives induce lamotrigine metabolism: evidence from a double-blind, placebo-controlled trial. Epilepsia 2007;48: Acknowledgments This study was supported by an investigator-initiated grant from Bayer Healthcare Pharmaceuticals, Inc.